. .

Probabilistic Segmentation in Virtual Reality

Probabilistic Segmentation in Virtual Reality

By In PhD proposals 2018 On February 13, 2019

Project: Probabilistic Segmentation in Virtual Reality

Laboratory: Decision and Bayesian Computation

Affiliation: Institut Pasteur/CNRS UMR 3571
Address: 25 rue du docteur roux
Website: https://research.pasteur.fr/en/team/decision-and-bayesian-computation/
E-mail: jean-baptiste.masson@pasteur.fr

LAB Director
Name: jean-baptiste masson
Phone number: 0144389267
E-mail: jean-baptiste.masson@pasteur.fr

Name: jean-baptiste masson
Phone number: 0144389267
E-mail: jean-baptiste.masson@pasteur.fr

Subject Keywords: Virtual Reality, One Shot Learning, Breast Cancer, Probabilistic Induction
Summary of lab’s interests: In the Lab, we question the computational basis of environment exploration and decision making from microscopic to macroscopic dynamics. We approach these topics through collaborative experiments and through theoretical approaches combining Physical based modeling, Bayesian Inference and Statistical Theory.

The lab currently focuses on 3 topics:

• Bayesian inference of single biomolecule dynamics with physicists definition of Random Walks ( https://goo.gl/4UH7NZ )
• Modeling the link between neural architecture and behavior of drosophila larva ( https://goo.gl/28TfNC )
• Data treatment leveraging Virtual Reality (https://goo.gl/dnNueu )

Project summary: The DIVA (Data Integration and Visualization in Virtual and Augmented Environments) project aims to create environments for data treatment that leverages virtual reality (VR), human-data interaction and automated algorithms. It is a joint initiative between the Institut Pasteur and the Institut Curie. The proposed PhD fellowship involves developing machine-aided segmentation approaches to volumetric imaging data. Key to our segmentation approach is the relation between user intervention and data treatment. It is tailored to handle data in very limited quantities, possibly very noisy, and to allow rapid and robust learning of the required procedures to extract relevant information from data.

Development of the proposed probabilistic segmentation tool will be based on an in-house software platform that is designed to load any type of image stack (e.g. microscopy image or MRI/CT scan) into VR environments. The approach will be focused on complex breast cancer MRI/CT scans where segmentation is neither obvious nor is there consensus on among radiologists where the limits of the tumours are. More information on applications of the DIVA project is available in https://goo.gl/dnNueu

Interdisciplinary aspect of the project: This project joins a new field, the use of Virtual Reality in scientific research and biomedical application, to the ever-growing field of probabilistic machine learning with applications in biology and biomedicine. The project involves developing within a new software platform, DIVA that generates automatically representation of any data in VR, new tools to interact with data and new numerical treatments on data while the representation is being actively modified.
Funding: The lab does not have prior funding for this PhD project (at this time) but 2 software engineers, 1 postdoctoral fellow are working on the project and it is being prepared to become a company.